How do I design a device to automatically switch to the backup battery?

Question

I've been told that many kinds of batteries work best if they are used until they are completely drained, and then recharged.

(Edit: The "memory effect myth" is quite widespread. Batteries work just as well if they are "topped up" every time.)

Right now I design devices the standard way: try to use the battery power as long as possible, and then when there is no power left it doesn't work at all.

If there was some simple way of implementing the following, I might use it from now on:

• Two independent batteries (or more)
• Once you start using one battery, continue to use it until it is completely drained reaches the manufacturer-recommended minimum voltage.
• When the "in use" battery is completely dead, switch to using the next battery. Presumably you use some technique similar to the "Switch between 5V power supplies?" question.
• Once you switch to the last battery, go into some sort of low-power mode so it can still do the important things in "limp mode", but the user is notified to plug it into a charger as soon as possible.
• After plugging into the charger, go back to standard mode -- but if disconnected from the charger before the batteries are fully charged, go back to limp mode.
• (optional) When plugged into the charger, only charge the drained battery (or batteries) and keep them topped off; leave the one "in use" battery alone.
• (optional) keep track of precisely how much total energy could be extracted from each particular battery the most recent time it was fully drained. Use that number (perhaps modified using Peukert's law) to give extremely accurate estimates of future run-time.

Why don't all devices do this?

• cell phones: first battery: talk like crazy. last battery: only emergency calls.
• laptops: last battery: throttle back to a slow speed that is adequate for looking up static documents
• handheld GPS: last battery: try to reduce energy by updating the screen less often, dimming the backlight, etc.

Answer 1

Why don't all devices use this? It adds cost and complexity. Is their any other reason for not doing something?

Seriously, I'd say that there are plenty of options and implementations for this. Having two equal batteries doesn't make much sense, so often the second is used for emergency or limp-home power. For instance, your PC has a RAM retaining battery on the motherboard for when you loose power. A laptop often gives a "Low battery" warning, at which time you're welcome to reduce power however you can.

I think that your statement that 'batteries work best if they are used until they are completely drained, and then recharged.' is a little broad. This is more the case for Nickel-based (NiCd and, to a lesser extent, NiMH) chemistries. Lithium Ion cells don't suffer this memory problem. In fact, their lifetime improves if you avoid deep discharges. See this page from BatteryUniversity.com for reference.

There are a couple of options for doing more intelligent power management in your own devices.

The simplest is an ORing diode on the power supply. If all you want is a hot-swappable power supply and you have a bit of leeway for your inputs, you can connect backup battery to the anode of a diode, and connect the cathode to your main battery. When the voltage of the main battery dips to 0.7V less than your backup (Or is removed), the other battery kicks in. Be careful of leakage current into the backup battery, it might overcharge it.

Alternatively, you can use a power mux IC like the TPS110. This lets you select your input independently (or dependently, if you prefer) of the input voltages, instead of always using the higher supply.

Finally, Linear Technology incorporates what they call "PowerPath" controllers into their battery charging ICs. I've used their LTC4011 which seamlessly transitions between battery and external power, and charges the battery while running off of the external power.

Answer 2

Nickel Cadmium cells were the infamous batteries which could build up a "memory" effect if not periodically cycled.

Newer Lithium-Ion rechargeable cells are considered immune from the memory effect. In fact, its best not to fully discharge them (stored in a fully discharged state can reduce lifetime), hence most electronics enter a "death" state while they can still monitor energy but refuse to enter a high power state.

Many devices, included laptops, can very accurately monitor battery state via their charger's coloumb counter and voltage of the cell(s). This allows all of the above behavior without the need and cost of duplicate charge circuits and batteries.

Answer 3

Completely discharging you battery is bad for all batteries (well, maybe not lithium-hydrogen batteries, but if you have those in your laptop, you know what you're doing anyways).

The practice originates with NiCad batteries, due to a confluence of unfortunate design decisions. The common-knowledge belief that NiCads loose capacity if not fully cycled is false.
Basically, the shallow-cycling of NiCad batteries results in a depression of the discharge curve, resulting in many electronic devices which use NiCads incorrectly reporting the battery as being empty. Note that the capacity of the cell is not significantly reduced, the cell voltage is merely depressed by a small amount. The fact that the major portion of the NiCad discharge curve is very flat leads to a small change in overall cell voltage causing a large change in the battery "Charge State" readout.

Note that the memory effect occurs only with NiCads, and was never actually even present in any other cell chemistries. The conflation of battery types, and naive assumption that all batteries are the same is what has lead to the belief about all batteries needing periodic cycling.

It is probably more accurate to think about batteries as being able to store and release a certain amount of energy, rather than being able to withstand a certain number of cycles. Whether this energy is release in 500 half-cycles or 1000 quarter-cycles is largely not relevant (An increased depth of discharge will actually reduce the total overall energy capability of batteries, though not significantly unless the batteries is discharged completely).

The manufacturer's that state that you should fully discharge and recharge your devices battery are stating this solely so that the internal battery charge measuring electronics can properly calibrate themselves. Basically, over time, the charge-state measurement will drift due to slight variance in the efficiency of charging and discharging the battery (basically, it's an integrator). A single full-cycle lets the electronics accurately measure the battery capacity without having to guess about the battery condition. The fact that if not fully cycled, battery measurement electronics will inaccurately report the battery state has led to the perseverance of the Memory Effect myth. The battery has not actually changed, the electronics are merely reporting an incorrect charge state.

As far as I know, the biggest cause of gradually reducing battery life in electronics is time. Lithium batteries actually have a shelf-life rated in a few years, whether they are used or not. Proper storage practices can extend this shelf life, but they will decay over time, no matter what use you put them through.

Deep cycling lithiums is actually quite bad for them, actually. Don't do it unless you have to (try to stay above 20% SOC).

Note: I am skipping a few things, like the issues with whisker growth in nicads. See the below link on NiCad for further reading.

See:
http://en.wikipedia.org/wiki/Memory_effect
http://en.wikipedia.org/wiki/Nicad